op5 Monitor 5.2 is about to be released to the public! By the time you read this it’s probably already out there 
.
This release is mostly about our new method for distributed, redundant and load balanced setups but we have managed to squeeze a lot of other nice features into the release as well. op5 Monitor users will get all this when upgrading to the 5.2 version and a lot of the features are available for Ninja and Merlin users as well.
Distributed, redundant and load balanced setups.
op5 Monitor 5.2 introduce Merlin as the base for distributed and load balanced setups. Merlin has been around for over a year as the backend for the new web interface introduces in op5 Monitor 5.0. We now take Merlin to the next level. To fully describe the possibilities available we need to define three different kinds of roles that a op5 Monitor server can take.
Master
A master server have one or more worker nodes, called pollers (described below) that do some or all of its work. The result of all checks performed by a poller is sent back to the master which mean the master server is in total control and contain all current as well as historical monitoring data. The master is also the place where all configuration is done and it sends relevant part of the configuration to his pollers.
Poller
A poller node gets his monitoring tasks from a master server. The poller perform the monitoring job assigned to him and send all results back to the master server. The poller is also responsible for sending out notifications regarding the objects assigned to him.
Peer
Two or more nodes can share the same tasks, we call them peers. Both masters and pollers can can be peered. Peers share the exact same configuration and automatically divide the work between each other. If one peer go offline the other(s) take over the failed nodes work.
Typical setup
You can configure your distributed/load balanced setup in a number of different way. This is not the place for full documentation of the different setups but we will look at two common examples.
Load balanced setup (peers)
Two or more op5 Monitor servers share the job of monitoring the environment. Configuration is distributed automatically between the nodes, IE you can make configuration changes on any node, the changes are automatically distributed to all nodes. If one of the peers fail the others automatically divide the additional workload between them.
This is the setup we recommend for users looking for redundancy or load balancing. For example, a single op5 Monitor setup can be expanded to a peered setup for enhanced security, failover and redundancy.
Distributed setup
In a distributed setup a master server distribute the work to one or more poller servers. All configuration changes are made at the master and automatically distributed to the pollers. The poller do the monitoring and feed results back to the master. If a poller fail the master try to taker over the work. The pollers are also responsible for sending out notifications for the objects they are monitoring. For enhanced security and redundancy the setup described here can be expanded with peers at all positions. You can, and in many cases we recommend that you, have dual peered masters to reduce the risk of loosing monitoring in case of hardware failure, network outage or other problems.
Web interface improvements
Even though this release is a lot about the distributed and load balanced features there a number of interesting new features related to interface usability.
Schedule Recurring Downtime
One of the main new features is the possibility to schedule recurring downtime. This has previously only been possible to achieve through third party addons but will now be available in Ninja. The link for this will only be available to users that has the authorized_for_all_hosts directive set.
Schedule Alert Summary Reports
It is now possible to schedule Alert Summary Reports the same way as with Availability and SLA reports. Schedule periods are also the same: Daily, Weekly and Monthly. This way you can now push the correct report to yourself or anyone else in your organization
Multiple commands
op5 Monitor 5.2 introduce the possibility to issue commands for several objects at once. This could, for instance, be used when you would like to acknowledge a bunch of services. In previous versions you would have to click to get to extinfo, select the command, enter some information in a form, submit and then go back and repeat the same steps for each host or service. With op5 Monitor 5.2, you just select the hosts/services and then the command to issue. This feature will drastically reduce the amount of time needed for those everyday tasks.
Graph and comment popup
Just like in previous versions there are icons related to hosts/services that indicate the presence of comments and performance data graphs. Now, when hovering these icons you get a popup previewing the graph of the last 24 hours or the comments associated with the host or service.
Paginated host/servicegroup pages and search result page
One thing that has been quite annoying in op5 Monitor 5.0 is the lack of pagination in search results. This has been fixed and while we are at it we added pagination to the host/service-group pages as well.
Customize widgets
One thing our customers love about op5 Monitor 5.0 is its flexibility, especially the possibilities to customize the tactical overview page. For this reason we introduce the possibility to change the thresholds for the ‘Network health’ widget and the background colors of the ‘Unhandled problems’ widget to give the user an even more customized view.
My account, per user configuration
Since the standard settings controlling the behavior and the look and feel of op5 Monitor might not be perfect for all users we introduce ‘My account’. This is a settings page were you control some of the interface settings. In this version each user can control:
- Pagination limit and step
- Skin (default and classic currently available)
- Keyboard command on/off. Keyboard commands can be used to search, pause page refresh and step forward/backward in paginated result sets.
New widget – Node status.
The ‘Node status’ widget has been introduced in op5 Monitor 5.2. Check the ‘Settings’ icon on the tactical overview if you want to add this widget. The widget shows the number of hosts/services monitored and the host/service check latency. This is very interesting for all op5 Monitor installations but the brilliant thing about this widget is that it shows you this information for all op5 Monitor servers in a distributed or load balanced setup!
User manual included.
The user manual is now included as a part of the product install. Follow the ‘op5 Monitor manual’ link located in the ‘About’ section in the left side menu.
Configuration interface
A number of new features has been aded to the web configuration interface.
Snmp v3 support
The ‘Scan host for SNMP interfaces’ function now support snmp v3. The scan do, however, require new check commands which are included but needs to be imported into the running configuration. To import commands you browse to ‘Configuration’ → ‘Commands’ → ‘Check Command Import’. Import all check commands related to check_traffic, check_errors and check_ifoperstatus.
Copy = clone to self
The clone function has been very popular, now we add ‘Copy’ which basically is clone but to the same object. This is very useful when you, for example, have a service monitoring the c: partition and want to add monitoring on the d: partition.
Plugins
A number of plugins has been bug fixed and we also added a couple of new plugins and new features to existing plugins. Highlight changes are
- improved performance in check_esx3
- adding check_email_delivery with support for SMTP with TLS auth and IMAP with SSL encryption.
